Introduction

In recent years, internationalization has become one of the most important issues in higher education in Taiwan. For example, the measure of internationalization has been included in the first university evaluation system as one of the six indicators that have been used to assess the quality of universities in Taiwan since 2005. Central to the effort of internationalizing Taiwan’s higher education has been a widespread promotion of the adoption of English as the medium of instruction (EMI), with many universities seeking to recruit “Western professors” to strengthen their efforts to construct advanced campuses (Chen 2010; Huang 2005). Table 1 indicates the number of foreign teachers who were employed by Taiwan’s universities and colleges from 2009 to 2012 (Department of Statistics of the Ministry of Education 2012). Based on Table 1, we can find a rising trend in the number of foreign teachers. Furthermore, the numbers in the table only included full-time foreign teachers; these numbers would be much larger if part-time teachers were included. With the spread of English as a global language (Chang 2010; Coleman 2006; Evans and Green 2007), most foreign teachers in Taiwan are English speakers, unless they teach another language as their profession, such as Japanese, Russian, or French (Department of Statistics of the Ministry of Education 2012).

Table 1 Summary of foreign faculty employed in Taiwan’s Universities (2009–2012)
Full size table

Traditionally, foreign teachers are mostly hired by the department of foreign Languages in universities in Taiwan in response to the need for English teaching or teaching of other foreign languages. In recent years, the influence of internationalization on higher education has encouraged other departments, such as sciences, mathematics, art, and even humanities and social sciences, to recruit foreign teachers and adopt some EMI courses in these departments. Generally, humanities- or social sciences-related subjects need refined expression or place emphasis on teacher–student interaction. Taiwanese students who speak Mandarin as their mother language, however, are not familiar with or even fear English to some extent. Moreover, departments with foreign teachers are usually concerned with small student numbers in EMI courses or consider the foreign teachers’ academic specialties and, thus, make these courses compulsory. Therefore, the result is that students, regardless of their English proficiency level, are required to take some EMI courses taught by these foreign teachers.

Although both the Ministry of Education and universities in Taiwan have promoted the internationalization of higher education to attract more international students to come to study in Taiwan and to increase students’ interest and motivation for learning the English language in the EMI courses, the higher education field needs to place more attention on the quality of students’ learning experience in these courses. What are the classroom learning experiences in foreign teachers’ EMI classes among Taiwanese students, who are non-native English speakers? Particularly, students who are majoring in humanities and social sciences may confront greater challenges in EMI courses, because these subjects require a higher level of language expression and teacher–student interaction. Therefore, it is worth identifying the similarities and differences between these students’ classroom experiences in foreign-faculty classes and local-faculty classes.

In a foreign teacher’s class, the level of the students’ proficiency in language is one of the most significant factors that influence the effects of classroom interaction with their teacher, classmates, and even themselves (Wilson 1991). In Taiwan, students in public universities typically perform better academically because public universities can usually enroll students with higher college entrance examination scores only (Yung 1999). Taiwanese students’ level of English is generally not good (Lee 2004; Lu 2009), and the level of private university students’ English is significantly lower than public university students’ (Chang 2006). Thus, we also need to examine whether students’ classroom learning experiences in foreign teachers’ classes are different between public and private universities. If there are differences, do these differences also appear in the local teachers’ classrooms?

Literature review

The theory of classroom experience

For university students, the classroom learning experience is very important because social and academic integration occurs in the classroom, which is the focal point of the higher educational structure (Demaris and Kritsonis 2008). Fassinger (1995) defined a classroom as a group with norms created by faculty and students. He pointed out that classmate traits, teacher traits, and student traits should be considered in discussing classroom experience because each of these factors influences their learning processes and results.

Regarding classmate traits, Fassinger (1995) suggested that classmate support and reciprocal expectations play a crucial role in classroom interaction. Compared with teacher–student interaction, student–student interaction is referred to as an “informal structure” (Terenzini et al. 1999; Weaver and Qi 2005). Nevertheless, for some students, peers’ informal interaction influences their learning even more than the formal structure (Alderman 2008). Students’ cooperation with each other in discussing and solving problems creates a positive atmosphere. However, there are sometimes negative effects of interaction, such as pressure from competing with peers, some students dominating other students, and observing of students’ admiration for classmates who are role models (Fassinger 1995).

Regarding teacher traits, following Hargreaves’ (1975) use of the symbolic interaction theory to analyze the relationship between teachers and students in the classroom, teacher–student interaction has become the basic concept for studying the classroom experience. Teaching is an extremely elaborate interaction process, and the teacher plays a crucial role in it. Fassinger (1995) proposed and emphasized the importance of the teachers’ supportiveness and approachability to students. Supportiveness includes teachers’ listening to and respecting the students’ different opinions, taking advice from students, and providing them with enough time to answer questions. Similar to supportiveness, the teacher’s approachability involves understanding the students’ feelings, sometimes tolerating student disagreement, and making students feel neither stress nor authoritarianism.

With respect to student traits, previous research has mainly focused on before class (study preparedness), during class (classroom participation), and after class (learning comprehension and confidence). Preparedness means that students can be expected to preview or review material to prepare for the next class or complete their homework before the next class (Fassinger 1995). Tinto (1997) found that the more the students prepare, the better the learning experience that they have and the better the outcomes they achieve from their classroom experience. Inadequate preparation may also affect the willingness of students to attend or participate in class (Loftin et al. 2010). Another student trait is classroom participation. Weaver and Qi (2005) pointed out that classroom participation is typically understood as students’ remarks or questions directed toward the instructor and as taking place within the confines of the classroom. Most students are reluctant to participate too overtly in class so as not to create displeasure among their peers. Nevertheless, some students may wish to communicate more for the purpose of earning a better grade. Another form of class participation includes nonverbal gestures, which express agreement or disagreement with faculty through body language. The third student trait is comprehension, consisting of the understanding of content covered in class and understanding the instructor’s questions (Fassinger 1995).

To summarize, a university classroom, like any other workplace, is a social organization where power is asserted, tasks are assigned and negotiated, and work is accomplished through the interplay of formal and informal social structures (Weaver and Qi 2005). Different classroom environments may lead to different students’ learning outcomes. This study focuses on the similarities and differences of students’ classroom experiences in foreign- and local-faculty classes. Specifically, the student traits, teacher traits, and the class traits constitute the scope of our discussion.

Classroom experience in foreign-faculty classes

What problems occur when students in EMI encounter foreign teachers who use a language that differs from the students’ native language? Wilson (1991) pointed out that language, culture, and pedagogy are the most frequent problems in a foreign-faculty class; other research has also shown similar results (e.g., Gorbunova 2003; Kamhi-Stein 2000). However, the subjects of these studies are American students; additionally, their foreign teachers’ or teaching assistants’ native language is not English, but they must use English to teach in class. This is very different from the problems that are confronted in Taiwan.

Another stream of the literature focuses on the classroom experience of students with English as their second language (ESL) in foreign university classrooms. This is similar to what we are discussing in this paper. However, ESL students are typically a minority in the classrooms, so class traits, such as peers’ interaction, are different. Poor English language proficiency has been found to be the single largest barrier for ESL students’ classroom experience (Olaniran 1996). This is also supported by Poyrazli et al. (2004), who indicated that English proficiency uniquely contributes to the variance in students’ acculturative stress. Good English proficiency facilitates the students’ successful adaptation to the new learning environment. Zhou et al. (2005) reported that ESL students considered themselves to be primarily listeners and explained that they lack the confidence to speak up in class. Liu and Littlewood (1997) also found that ESL students’ most frequent and preferred classroom experience was listening to the teacher. Specific to Asian students, Sauman (1999) indicated that they were reluctant to appear critical in class and were concerned about being perceived as polite and respectful to their instructor.

Whether or not the above-mentioned circumstances are present, as long as language and cultural background differences between faculty and students exist, students’ classroom experiences can be significantly affected. However, only a few studies (e.g., Chang 2010; Hsieh and Kang 2007; Wu 2006) have been conducted to investigate undergraduate students’ classroom experiences of EMI courses, and to the best of our knowledge, no studies have yet been conducted on EMI courses taught by foreign faculty members in Taiwan. Hsieh and Kang examined the influences of EMI on courses at National Taiwan University. In their study, a local professor taught a civil engineering graphics course in two different semesters to two different groups of undergraduate students. One group (n = 47) received instruction in Mandarin; the other group (n = 19) was taught the same course in English in the following semester. They found that there was no obvious difference between these two groups in terms of grades. Nevertheless, students in the EMI class tended to show a more positive learning attitude, and they also felt that their proficiency in English skills had improved over the semester.

Similar results were also found in Wu’s (2006) study of reactions toward EMI from 28 engineering graduate students at Chung Hua University. The result of his study indicated that most students thought that EMI helped them to improve their English proficiency, even though they did not have a good command of English. His students reported that EMI created greater difficulties in understanding course content and inhibited them from expressing themselves freely in the class.

Recently, Chang (2010) investigated 370 undergraduate students who took EMI courses in two departments from three colleges (the College of Engineering, the College of Management, and the College of Informatics) at a private university in northern Taiwan. Chang found that most of the subjects did not show negative attitudes toward the courses, probably due to their professors’ varied efforts to reduce their anxiety level in the classroom. Moreover, while the effects of EMI on the learning of subject content remain unclear, most of the students surveyed agreed that EMI helped them to improve their English language proficiency, especially in terms of listening.

Although Hsieh and Kang (2007), Wu (2006), and Chang (2010) reported similar results in their studies, the EMI courses that they studied, again, were taught by local Taiwanese teachers, not foreign teachers. Additionally, the EMI courses in their studies were engineering, business, or information science courses, which differ greatly from humanities or social sciences courses in terms of students’ classroom experience. Thus, Taiwanese students’ classroom experience in the EMI humanities and social sciences classes taught by foreign teachers remains unclear. Would these experiences differ from those in local-faculty classes? How would students’ perceptions of the classroom experience of the foreign faculty and the perceptions of classroom experience of the local faculty differ?

Classroom experience in public and private universities

Unlike the situation in the western world, with the help of a unified national college entrance examination, public universities in Taiwan usually get the “better” students. They can also recruit qualified faculty members more easily, establish better research and teaching facilities, and offer better training to students (Chang et al. 2011). They can also afford to do so while simultaneously charging low student fees in comparison with private universities, because of government subsidies (Yung 1999). Therefore, there are sufficient grounds to consider the university’s type (public or private) when discussing university students’ classroom learning experience in Taiwan.

According to Pascarella and Terenzini (2005), although the type of university has a statistically significant effect on its students, the effect is not very practical for teaching and learning in higher education. However, their findings suggest that private university students in the United States perform better than public university students in terms of learning outcomes (Astin 1993). Astin explained that the size of private universities was typically smaller than public universities and, thus, students had better teacher–student interactions in private university classes. Astin’s research setting, however, is different from Taiwan. His sample is from the private elite universities in the United States, while private universities in Taiwan typically have hundreds of thousands of students with relatively lower academic achievement compared to their public counterparts. In Taiwan, both public and private universities are attempting to adapt to the trend toward internationalization, in terms of hiring more foreign faculty to teach. However, in public universities, the students’ English level is usually better. Therefore, we believe that it is necessary to compare the students’ classroom experience in foreign-faculty classes between private and public universities, to propose a reference to improve the quality of classroom teaching.

Method

Survey process and sample

As discussed in the Introduction, this research focuses on the fields of humanities and social sciences since courses taught in both domains, generally, are expected to be interactive regarding classroom instruction, hence requiring higher level of language proficiency. Therefore, we selected students majored in humanities or social sciences from 14 universities in Taiwan, who, during 2008–2010, had taken at least one EMI subject course taught by a foreign teacher. That is, the participants in this study had been enrolled in both the foreign-faculty EMI courses and the local-faculty courses taught in Chinese.

The mail goal of this study was to compare students’ classroom learning experience between foreign-faculty and local-faculty classes at public and private universities in Taiwan. Therefore, we employed the split-plots ANOVA design, where type of university is an independent variable, and the type of class (foreign faculty vs. local faculty) is a dependent variable. That is, we surveyed the same students, who had taken both types of classes, to test for their perceptions of classroom learning experience between foreign-faculty classes and local-faculty classes across public and private universities. Thus, instead of the simple random sampling method, a multistage cluster sampling method was appropriate for our study. We conducted multistage cluster sampling: Universities were chosen for the first stage, and classes were randomly selected for the second stage.

A total of 14 general universities that offered EMI courses taught by foreign faculty agreed to participate in our study, including three public and two private universities in northern Taiwan, two public and two private ones in central Taiwan, two public and one private university in southern Taiwan, and one public and one private university in eastern Taiwan. The analysis was finally based on useable responses from 97 % (N = 714) of the invited sample. The sample varied in its range of demographic factors, as shown in Table 2.

Table 2 Summary of sample demographics (N = 714)
Full size table

Instrument

The instruments used in this study were the Students’ Perceptions of Classroom Experience with Local Faculty scale (SPCE-LF), and the Students’ Perceptions of Classroom Experience with Foreign Faculty scale (SPCE-FF). The items of these two scales were derived from the previous literature (e.g., Fassinger 1995) and a focus group interview of 13 Taiwanese university students. The 13 university students in the focus interview group were from one of the 14 selected universities where the authors have been teaching up to now. These students had taken at least one EMI subject course taught by a foreign teacher before the interview. Some contents of items in the instruments used in this study were derived from the interview with these students. The mean concept, “classroom experience,” was divided into three dimensions, including class traits, faculty traits, and student traits, according to Fassinger’s (1995) study. Class traits mainly referred to peers’ interaction, including peer supportiveness and peer pressure; faculty traits consisted of teacher supportiveness and teacher approachability; student traits included study preparedness, class participation, and learning comprehensiveness. Each of the seven subscales consisted of three items, and the questionnaire was composed of 21 five-point Likert items, with anchors ranging from strongly disagree (1) to strongly agree (5).

To compare the similarities and differences in students’ experience between local- and foreign-faculty classes, the SPCE-LF edition and the SPCE-FF edition were created in the same form, with the exception that each item of the SPCE-LF referred to local-faculty classes and the SPCE-FF referred to foreign-faculty classes (see Table 3). Participants’ background information was also collected by the questionnaire. Table 3 shows the construct validities and reliabilities of the SPCE-LF. Except for Item 6 (factor loading = .33), factor loadings for the SPCE-LF designed to measure each factor were between .64 and .84. The seven factors accounted for 68.27 % of the total variance. Except for peer pressure, the reliabilities of most of the scales were considered acceptable; the factor of peer pressure, whose Cronbach’s α was .48, was mainly caused by Item 6.

Table 3 Construct validities and reliabilities of the SPCE-LF of the SPCE-FF
Full size table

The construct validities and reliabilities of the SPCE-FF are also indicated in Table 3. Similar to those of SPCE-LF, the factor loading of Item 6 was also small. Other factor loadings for the SPCE-FF designed to measure each factor were consistently large, between .63 and .84. The total variance explained by the seven factors was 70.82 %.

Analytic strategy

The means and standard deviations were calculated for the perceived classroom experience variables in local- and foreign-faculty classes across the public and private universities. Then, we applied a 2 × 2 split-plots factorial research design to analyze the significance of the factors’ main and interaction effects on the students’ classroom learning experiences.

Results and discussion

Table 4 presents the means and standard deviations of the dimension scores of public and private universities. The ranges of the means of all of the dimensions indicated that students’ perception of classroom experiences in either local or foreign classes was higher than moderate for both types of universities.

Table 4 Summary of student perception of classroom experience
Full size table

Peer traits

Table 5 shows the summary of the analysis of variance across university type (private vs. public) and faculty type (local faculty vs. foreign faculty) for peer supportiveness and peer pressure. We find that university type and faculty type had an interaction on students’ perception of peer supportiveness, F (1, 710) = 12.99, p < .001, which is also displayed in Fig. 1. The figure and table also indicate that, in foreign-faculty classes, private university students obtained a significantly higher score on peer supportiveness than did the public university students; private university students’ peer supportiveness score in foreign-faculty classes was also higher than in local-faculty classes. However, there was no significant difference between public and private universities on peer supportiveness in local faculty’s classes.

Table 5 ANOVA of class traits across university type and faculty class
Full size table
Fig. 1
figure 1

University type by faculty class on peer supportiveness

Full size image

In regards to peer pressure, both types of faculty type and university type had a significant effect on peer pressure, and they also had an interaction, F (1, 712) = 14.15, p < .001. As shown in Fig. 2, for both types of universities, students felt more pressure from their peers in foreign-faculty classes than in local-faculty classes. In addition, students of public universities felt more peer pressure than private university students in local-faculty classes, whereas there was no distinction in foreign-faculty classes.

Fig. 2
figure 2

University type by faculty class on peer pressure

Full size image

Faculty traits

As shown in Table 6, teacher supportiveness, F (1, 712) = 63.71, p < .001, and approachability, F (1, 712) = 65.88, p < .001, were higher in foreign-faculty classes than those in local-faculty classes, for both types of universities. The results are presented in Figs. 3 and 4. They suggested that, compared with Taiwanese teachers, foreign teachers’ pedagogy encourages disagreement and discussion, as well as respects students’ opinions. In addition, Table 6 and Fig. 4 indicate that private universities obtained a significant higher teacher approachability score than did public universities, F (1, 712) = 3.99, p < .05. However, no significant factorial interaction effect was found for the two teacher traits.

Table 6 ANOVA of teacher traits across university type and faculty class
Full size table
Fig. 3
figure 3

University type by faculty class on teacher supportiveness

Full size image
Fig. 4
figure 4

University type by faculty class on teacher approachability

Full size image

Student traits

In terms of study preparedness, faculty type and university type produced an interaction effect, F (1, 712) = 4.24, p < .05, on the students’ preparedness for class, as shown in Table 7 and Fig. 5. Study preparedness was higher in foreign-faculty classes than local-faculty classes in private universities, according to the post hoc main effect analysis.

Table 7 ANOVA of student traits across university type and faculty class
Full size table
Fig. 5
figure 5

University type by faculty class on study preparedness

Full size image

Interaction effects of faculty type and university type were also found for class participation, F (1, 711) = 4.29, p < .05, and learning comprehensiveness, F (1, 711) = 4.16, p < .05 (see Table 7). As Fig. 6 shows, for both public and private universities, students’ class participation in local-faculty classes was significantly higher than in foreign-faculty classes. For local-faculty classes, private university’s class participation was higher than for public universities. However, in foreign-faculty classes, there was no significant difference between class participation in public and private universities.

Fig. 6
figure 6

University type by faculty class on class participation

Full size image

Regarding learning comprehensiveness, we find from Table 7 and Fig. 7 that both public and private universities obtained higher scores in local-faculty classes. Moreover, public universities got a significantly higher score in foreign-faculty classes than private universities; however, no distinctness was found in local-faculty classes. This was probably because public university students’ English proficiency was better than private university students in Taiwan.

Fig. 7
figure 7

University type by faculty class on learning comprehension

Full size image

Discussion

The results of these analyses reveal some important points. Firstly, both faculty type (local or foreign) and university type (public or private) influence students’ classroom learning experience. In Taiwan, university type implies students’ level of academic and English proficiency, and faculty type reflects language, culture, and pedagogical differences. In other words, students’ level of academic and English proficiency, faculty’s cultural background, as well as the language and pedagogy that they use, influence students’ classroom learning experiences. This finding corresponds to Blumer’s (1969) symbolic interactionism theory. Students and teachers are the two leading actors in the classroom. If either of them has different backgrounds, different classroom interactions and learning experiences may arise, i.e., interaction is produced by both sides. When confronted by a foreign teacher, the students will interact differently compared to their interaction with a local teacher, and this difference may vary according to the students’ abilities and prior experiences.

Secondly, the interactions of these two factors are found in class traits. In terms of peer supportiveness, private university students receive higher peer support than public university students in foreign teachers’ classes; additionally, peer support in foreign teachers’ classes is higher than in local teachers’ classes for private university students. In other words, generally speaking, private university students with a lower English level have to support each other to interact successfully with the English-speaking teachers. Meanwhile, in foreign-faculty classes, students also feel pressure from their peers in both types of universities. This supports Wilson’s (1991) finding that language differences would impose pressure on the students whose native language differs from the teacher’s. Therefore, private university students would look for peer support and encouragement to succeed in the foreign-faculty classes. Another possible reason for higher peer support in foreign teacher’s classes is that foreign teachers typically encourage students to work together as a team (Hsiao 2008); this is especially the case in Taiwan’s private universities, where the students’ English proficiency level is not so high (Lee 2004; Lu 2009).

Thirdly, throughout the three aspects of classroom experience, teacher traits are the only category in which no interaction was found, although it is significantly influenced by both faculty type and university type. The reason for this is probably that foreign teachers’ supportiveness and approachability are higher than local teachers’ in both public and private universities. Impacted by Chinese traditional culture, both Taiwanese teachers and students believe that remaining silent in the classroom is polite and respectful to the teacher (Sauman 1999). Conversely, western culture is characterized by a more lively atmosphere in the classroom. Western teachers encourage and praise students, and accept diverse opinions in comparison with eastern teachers. In addition, private-university student-reported teacher approachability is significantly higher than that of public universities. This is probably due to the mission of most private universities in Taiwan being to provide a good teaching and learning environment, therefore presenting teaching quality and student services as their core competitive advantages (Chang et al. 2010). On the other hand, under the pressure of the scientific-achievement-oriented evaluation system, faculties in Taiwan’s public universities are required to carry heavy research loads and, thus, tend to pay less attention to teaching and students (Lu 2009).

Another explanation is that perhaps the students stereotype their western/foreign teachers and the local Taiwan teachers. Many university faculty development centers (FDCs) in Taiwan have helped their faculty improve student–teacher relationships under the MOE’s Teaching Excellence Projects since 2005; however, students may still think that western teachers are more willing to accept diverse opinions compared to eastern/Taiwanese teachers.

Last but not least, private university students prepare more in foreign teachers’ classes than in their local teachers’ classes. However, no such difference was found in public universities. A reasonable explanation for this is that their poor English proficiency motivates the private university students to prepare well for the class (Chang 2006). However, both public and private students in foreign teachers’ classrooms are reluctant to participate and find it difficult to comprehend class content because of poor English proficiency. This finding partly corresponds to Hsieh and Kang’s (2007), Wu’s (2006), and Chang’s (2010) research in Taiwan, and Olaniran’s (1996) and Poyrazli et al.’s (2004) work on international students in the United States, finding that poor English proficiency is the single largest barrier in academic success. Conversely, good English motivates them to participate in the classroom and succeed in the class more quickly.

Conclusions and suggestions

Conclusions

This study investigates university students’ perceptions of classroom experience in local-faculty and foreign-faculty classes, and their relationships to institutional type. Based on this empirical study in Taiwan, we draw the following conclusions from our research:

Students in both types of universities are reluctant to participate, experience pressure, and have difficulty comprehending course content in foreign teachers’ classes. However, they feel that foreign teachers are more supportive and approachable than local teachers, especially in private universities.

Compared with local-faculty classes, private university students prepare more and receive more peer support in foreign-faculty classes. They participate more and experience more peer support; however, they comprehend less of their course content than their public university counterparts in foreign-faculty classes.

Suggestions for university administrations

In recent years, vigorous internationalization has been taking place in Taiwan’s higher education and other Asian universities. For example, the Teaching Excellence Project, sponsored by the Ministry of Education of Taiwan in 2010, added English-taught degree courses as a new criterion (Ministry of Education 2010). We can predict that more and more EMI courses taught by foreign teachers will be offered in universities in the future, not only in Taiwan, but also in non-English-speaking countries in the region. In addition to continuing to expand the employment of foreign teachers and designing EMI courses, the Ministries of Education and universities in these non-English-speaking countries should pay more attention to students’ learning experience in foreign-faculty classes. According to the findings of our research, students’ learning experience in foreign teachers’ classrooms differ significantly from that in local teachers’ classrooms. In the foreign teachers’ classrooms, although the degree of preparation and interaction within peers and between the teacher and students are high, students’ participation and learning comprehension are both low. In addition to this, student pressure is also higher. Without taking steps to strengthen learning strategies and improve English proficiency, promoting internationalization may hurt the students long before it benefits them. Therefore, we make the following recommendations for universities in Taiwan and the non-English-speaking countries in the region.

First of all, we suggest that management regularly examine students’ learning experience. Universities should measure students’ experience and satisfaction in the classroom, especially in the foreign teachers’ classroom, to identify students’ difficulties and provide them with directional assistance.

Secondly, we suggest that local faculty take the initiative and encourage their students to speak out and ask questions in class and, moreover, respect their opinions. Taiwanese students have long been influenced by the traditional Chinese culture of respecting teachers; thus, students are the passive side of the student–teacher relationship, and there is often a fear of approaching their teachers. However, our findings show that students feel that foreign teachers are more approachable. In other words, as long as the local teachers are willing to make friends with their students, they can change this dynamic. In regards to foreign teachers, we would suggest that they pay more attention to students’ English proficiency level and make their form of expression more easily understood.

Thirdly, we suggest that universities strengthen the training of students’ English to improve their experience in foreign-faculty classes. Internationalization is based on the premise that language obstacles have been overcome. Universities are supposed to create an atmosphere of learning and using English on campus. This atmosphere should intensify students’ willingness to take foreign-faculty classes. However, students’ willingness should be respected, since they should have the right to decide whether or not to take foreign teachers’ classes. After all, mastering the English language is difficult, especially for private university students in Taiwan. Therefore, if we encourage students to take a foreign teacher’s class, we must also respect their decision and choices. The findings of the research show that students would like to support each other and become well prepared for foreign teachers’ courses; so, we should encourage students to establish study groups or book clubs particularly related to their foreign teachers’ courses. Their seniors and teachers can be invited to their club to share their experiences and to solve problems.

One more suggestion concerns the flexible credit system. More time and energy are required for a student to study in a non-native language-taught class; thus, we should consider giving foreign-faculty courses more credits. This would allow the student to spend more time learning the course material well.

The last suggestion is related to students’ cultural identity. The increasing use of English-language teaching in Asia raises the question of whether or not it will inevitably lead to the displacement and erosion of national languages and cultural identities. We suggest that, although EMI may be suitable in science and economic studies, it seems far less appropriate in courses on humanities, history, social sciences, and Asian culture. The latter subjects are closely linked to particular cultural meanings embedded within each language, which cannot easily be expressed in others, such as English.

Limitations and suggestions for future research

Some inevitable limitations of the present study deserve attention. Firstly, participant selection may limit the generalizability of the results. The participants were not simply randomly assigned to different instructor classes (i.e., foreign or local), nor were they randomly assigned to private versus public institutions. Specifically, the dependent variables, i.e., students’ classroom learning experiences, were based on their self-reports—with no control for students’ prior scores on the instrument. Thus, there is a possibility that the findings of this study reflect individual student characteristics and propensities rather than instructor class or institutional type. One must be cautious when extending conclusions based on the results of this study. Further research is needed to explore the effect of students’ characteristics on their classroom learning experience using randomized experiments with statistical control.

Secondly, the participants of this study are mainly from the humanities and social sciences departments in 14 of Taiwan’s public and private universities. Other discipline areas or polytechnic universities are not included in our study. Since students in the humanities and social sciences fields are mainly female, 77.4 % of our participants are female. Students’ classroom experience in EMI courses could be affected by gender and several other factors as well, such as English proficiency level, international experience, participation in overseas exchange programs, major (English major vs. non-English major), years in school, course type, etc. Therefore, future research might examine students’ gender, personal background, and other discipline areas in order to understand the effects of these factors on students’ learning experience in foreign-faculty and local-faculty classes.

Thirdly, the reliability and validity of our instrument still leave room for improvement. In particular, the reliability of peer pressure was low in both the SPCE-FF and SPCE-LF questionnaires. Future studies need to increase the reliability and validity of the questionnaires in order to enhance the quality of the research.

Last but not least, this study was based upon university students’ self-reports of their sense of classroom experience, rather than upon field observation in classrooms. Self-reports frequently produce social desirability bias, which interferes with the interpretation of average tendencies, as well as individual differences. Thus, for future research, we suggest that researchers observe foreign and local teachers’ classrooms for long-term observation, supplemented by interviews. After all, classroom experience is a cognitive, affective, and interpersonal process, which is full of complex interactions and roles. Therefore, in order to comprehensively understand students’ experiences and feelings in foreign and local teachers’ classes, in-depth interviews are necessary.